To the Editor: An article in the September 1999 News and Notes section quoted a report from the Bureau of Justice Statistics (BJS) indicating that in mid-1998 about 63 percent of mentally ill offenders (179,200 individuals) were in state prisons, not local jails (
1). New York State has just conducted one of the largest studies on the overlap between the criminal justice and mental health systems. That study found that the preponderance of persons treated for mental illness who are in correctional facilities are in local jails, not in state prisons.
In contrast to the New York State study, which used a longitudinal methodology that permitted examination of data over a five-year period, the Bureau of Justice Statistics used a census-based methodology that captured data at one point in time. A census-based methodology is not appropriate for determining the total volume of persons with a mental illness in prisons versus jails because it does not accurately measure the high volume of individuals who enter and exit local jails over a period of time. Jails tend to be short-stay facilities while prisons are generally long-stay facilities, making point-in-time comparisons problematic.
Although the BJS survey finding that more persons with mental illness are in prisons than in jails is accurate for a point in time, it misrepresents the actual volume of persons with mental illness who quickly pass through local jails. It also stigmatizes persons with mental illness by giving the appearance that most of those who are incarcerated have committed crimes serious enough to warrant a prison sentence. The longer-term New York State study, which included data on more than 117,000 recipients of mental health services between 1991 and 1995, makes it clear that most persons with mental illnesses who are incarcerated go to jail and not to prison. It found that female recipients of mental health services are 17.6 times more likely to go to jail than to prison, and that males are 8.9 times more likely to spend a night in jail than in prison.
Understanding the volume of persons with mental illness in jails versus prisons is important not only from a strong concern to reduce stigma and discrimination; it also has clear policy and program implications for correctional and mental health care administrators. Our findings show that persons with mental illness who are involved in the criminal justice system are likely to have committed less serious offenses. Most have short stays in local jails, not state prisons. Therefore, services to reach the majority must be directed at jails and community programs.
The challenge for local and state governments is to understand why some recipients of mental health services are involved with the criminal justice system and to develop and evaluate service packages to divert them from incarceration.
